9b

Introduction

  • Founders are passionate about their companies and want to maintain a central role, especially as CEO.

  • Empirical studies show mixed results of founder-CEOs on performance.

  • Morck et al. (1988) found a negative effect of founding family control on market valuation, especially for older firms.

  • New ventures may exceed founders' managerial capabilities.

  • Founders may prioritize personal satisfaction over economic benefits.

  • Fahlenbrach (2009) found that firms led by founder-CEOs outperform others in stock performance and market valuation.

  • There is a tension between profit and control motives in founder-CEO companies.

  • Founders care about financial performance and socioemotional wealth (Berrone et al., 2012).

    • Socioemotional wealth: non-financial aspects meeting affective needs like identity or vision.

  • Deciding whether to fire a CEO, especially a founder or someone related, after poor performance is critical.

  • This paper investigates CEO turnover, controlling for founders' roles.

  • CEO dismissal decisions reflect the quality of internal corporate governance.

  • The sensitivity of CEO turnover to poor stock performance is a signal of effective governance.

  • Empirical studies confirm a negative association between CEO turnover and stock/firm performance.

  • Warner et al. (1988) found an inverse relation between management change and share performance in the US.

  • Weisbach (1988) found a stronger association between stock returns/earning changes and CEO dismissal for companies with outsider-dominated boards.

  • Studies confirm this relationship in various countries (Franks et al. (2001) for the United Kingdom, Volpin (2002), Brunello et al. (2003) for Italy, Lausten (2002) for Denmark,Suchard et al. (2001) for Australia, or Kang and Shiv- dasani (1995) for Japan).

  • Most studies focus on developed countries.

  • This study explores the relation between forced CEO turnover and firm performance in a country with low-quality institutions and governance standards.

  • Focus is on how CEO turnover is affected by founders, including when a founder is the current CEO or a board member.

  • After growth, founders may take a passive role as a member of the executive or supervisory board.

  • Founders can still influence a firm by influencing CEO replacement decisions.

  • If founders on boards are effective monitors, there should be a stronger relation between firm performance and forced CEO turnover.

  • Whether an external CEO succeeds a founder is strongly determined by the founder’s opinion.

  • The current CEO is likely to execute the founder’s strategy if the founder maintains a monitoring role.

  • The likelihood of the current CEO being fired is less when the founder is monitoring the company.

  • The paper uses a hand-collected sample of 359 companies listed on the Warsaw Stock Exchanges (WSE) from 2000 to 2015.

  • This paper contributes to the literature on founder-CEO replacement.

  • Existing evidence on management turnover and performance in developing/emerging countries with low governance is sparse.

  • Campbell II and Keys (2002) examined corporate governance in South Korea and reported a negative relation between CEO turnover and firm performance.

  • Campbell II and Keys (2002) found top executive turnover unrelated to performance for leading chaebol firms.

  • They concluded that internal corporate governance is weaker in chaebols.

  • Results contrast with Jenter and Kanaan (2015) and Adams et al. (2009), who did not find that founder-CEOs are immune to dismissals due to poor performance.

  • This study provides additional evidence on the turnover-performance relation, focusing on the impact of CEO shareholdings on internal monitoring efforts.

  • Some argue that CEO equity holders may become entrenched (Morck et al., 1988).

  • The evidence on the role of equity in the turnover-performance relation is mixed.

  • This study finds contradictory results for outsider CEOs and founder-CEOs, attributing them to differences in values.

  • A founder-CEO may protect the company at the cost of personal benefits.

  • The study explores the role of founders as monitors and their impact on CEO turnover.

  • Morck et al. (1988) provided evidence of entrenchment of founder-executives.

  • Firms with founding family members in top management are less likely to experience complete turnover or be targets of hostile takeovers.

  • Lausten (2002) analyzed the role of the chairman of the board in monitoring top executives and its impact on CEO turnover.

  • Lausten (2002) found that CEO turnover increases if the chairman is an insider.

  • Bresser and Thiele (2008) showed that a firm’s current CEO is more likely to be dismissed following poor performance when a former CEO is the chairman the supervisory board.

  • This study presents new evidence on these aspects.

  • It analyzes the transition of former CEOs to the supervisory board.

  • Andres et al. (2014) reported that the announcement of a transition of a retiring CEO to the supervisory board is considered good news by the shareholders.

  • They found that executive compensation and compensation of the supervisory board members increases after CEO transition to the supervisory board.

  • Andres et al. (2014) did not find any negative effect of CEO transition on the long-run share or operating performance.

  • Bermig and Frick (2010) found evidence of a negative relation between the percentage of former managing board members on the supervisory board and the firm’s market value.

  • The literature shows ambiguous results concerning the effect of CEO transition to the supervisory board.

  • No existing studies analyze what determines the transmission of a former CEO to the supervisory board.

  • This study investigates the determinants of CEO turnover and the CEO’s transition to the supervisory board, with a focus on the role of firms’ founders.

  • The paper is organized as follows:

    • Section 2: Institutional background

    • Section 3: Sample, data, and research design

    • Section 4: Empirical findings and evaluation

    • Section 5: Conclusions

Institutional Background

  • The WSE re-opened in 1991 and developed through privatization and IPOs of private companies.

  • In Poland, listed domestic-owned private firms are mostly first generation, where the founder also still serves as the CEO.

  • These companies were mostly established after 1989 and are relatively young compared with their counterparts in industrial countries.

  • The first privately founded firm, Efekt, went public in 1993.

  • The IPO of this company was successful.

  • In 1994, only 2 out of 22 public offerings were private companies, almost all IPOs were related to already privately-owned firms in 1999 (Kowalewski et al., 2008).

  • Some founders exited after the IPO, while others used capital for expansion.

  • Not all listed domestic-owned firms are successful long-term.

  • In 2018, GetBack (founded in 2012, public in 2017) filed for bankruptcy protection and restructuring.

  • The company was controlled by private equity funds and had foreign investors but failed to redeem its bonds a year after going public.

  • The founder-CEO was dismissed days before the default, which was a surprise.

  • The company was recognized as the fastest growing on the WSE a few weeks earlier.

  • The failure revealed weak governance, allowing the founder-CEO to alter financial statements to hide financial losses.

  • There was increased leverage from public and private bond issues.

  • The IPO prospectus and financial statements were audited by a large company, but discrepancies were not discovered until it was too late.

  • The internal governance structure of Polish joint stock companies is composed of a general meeting and a supervisory board.

  • The board system is two-tier, meaning all members are non-executives and CEO duality is prohibited.

  • The management board is responsible for business conduct, restrained only by the supervisory board.

  • The supervisory board is independent, monitors managers’ performance, and can appoint or dismiss managers.

  • There is no legal requirement for education or expertise of supervisory board members.

  • Members of the supervisory board are elected at the general meeting, meaning major shareholders can vote for members who meet the criteria of independence and yet follow their interests.

  • Internal governance standards are relatively weaker than those of developed countries (Kowalewski, 2016).

  • The chairman of the GetBack supervisory board, recommended by private equity funds, declared unawareness of the firm’s difficulties before its default.

  • External governance mechanisms are also weak due to limited contestability to hostile takeovers and the absence of institutional investor activism.

  • Bonin and Wachtel (2003) state that stock markets in Central Europe leaped into existence before the institutional infrastructure was established.

  • Share listings did not guarantee transparent share registration, transfer of ownership, or absence of price manipulation.

  • The legal system and external/internal governance mechanisms are not strong enough to control managerial discretion, especially for firms’ founders.

  • Founders can create costly firing mechanisms.

  • CEO turnover is an important ingredient of corporate governance.

  • The WSE is appropriate to test corporate governance functioning, considering the strong position of founder-CEOs and Poland’s weak institutions.

Empirical Approach

  • This section describes the data and outlines the main estimation technique.

Data

  • The data set covers companies listed on the WSE during 2000–2015.

  • Focus is on forced CEO turnover as these dismissals represent governance actions.

  • Dates of CEO turnover announcements and stated reasons for departure were hand-collected from company statements and news articles.

  • Forced CEO turnovers are classified as those where the CEO was fired, forced out, or resigned due to pressure.

  • Cases where the reason for CEO turnover could not be classified were excluded, especially for CEOs over 65.

  • A forced CEO change occurred in 484 year-observations, approximately 15%.

  • Table 1 shows descriptive statistics with variable definitions.

  • Information was collected on who established the companies and whether the founders maintain an active role in firm governance.

  • The information on founders was encoded using multiple dummy variables.

    • F−CEOF-CEO and F−MGTF-MGT take the value one if the founder is the CEO or only a member of the management board, respectively, and zero otherwise.

  • As companies develop, founders may move from top management to a supervisory board.

    • This allows influence but less involvement in day-to-day decisions.

    • The dummy variables F−ChairF-Chair and F−NEDF-NED take the value of one if a founder is the chairman or a member of the supervisory board, respectively, and zero otherwise.

  • A founder can also be forced to resign or fired.

    • The information on CEO turnover is supplemented by encoding the cases where a founder is dismissed (F−CEOTurnoverF-CEO Turnover).

  • Governance data is supplemented with information on insider ownership, retrieved from Reuters.

  • Firms that are state-controlled were eliminated.

    • The number of forced CEOs in the sample declined to 426 year-observations.

    • In state-controlled companies, CEO dismissal is more likely determined by political motives.

    • The regression results confirmed this assumption.

    • The coefficient of the dummy for state ownership was positively and statistically related to CEO turnover.

    • The results are available upon request.

  • Firms’ founder ownership are controlled for (F−S’holderF-S’holder) using a dummy variable that takes the value of one if the founder holds shares in the company and zero otherwise.

  • Many companies listed on the WSE are relatively young.

    • Founder ownership is reported in 900 year-observations, over 28% of the sample.

  • Some argue that when the CEO is an important equity holder, the officer can become entrenched (Morck et al., 1988).

    • CEO ownership is additionally controlled for using a dummy variable CEO S’holder that takes the value one if a CEO holds shares in the firm and zero otherwise.

  • There is a strong overlap between founders and CEO shareholders.

    • In almost 90% of year-observations when a founder is a shareholder, the founder is also the CEO.

    • This confirms that a significant share of firms are first generation firms taken public by the founder-CEO.

  • Financial data for the companies was obtained by extracting balance sheet information and financial ratios from Reuters.

  • Both accounting-based and market-based performance variables are employed.

  • For an accounting-based measure, return on assets (ROA) is used, calculated as net income over total assets.

    • Net income is focused on rather than EBIT as the sample includes companies from the financial sector.

    • ROA is a better measure to compare companies across industries, unaffected by industry characteristics.

  • Stock return (Return) is used as a second measure, adjusted for dividends and splits.

  • The performance variables' implications are radically different in each case.

    • ROA reflects a tangible, balance-sheet effect on companies’ performance.

    • Market return is concerned more with market perceptions of future earnings and the value of the company.

  • Kaplan (1994) suggests that stock returns also reflect changes in discount rates, and accounting earnings may be more informative.

  • The empirical evidence is ambiguous on this issue.

  • Two measures for firm performance are used to ensure robustness.

  • Some correlation is expected between the two performance measures, but this might not always be the case.

  • With reference to the firm’s characteristics, three control variables are included:

    • market value to book value (Market to Book)

    • total debt to assets (Leverage)

    • natural log of total assets (Size)

  • Leverage may affect CEO turnover.

    • The debt ratio can improve performance by limiting managerial misbehavior.

    • Bank financing is important in Poland.

    • Institutions providing debt financing may play an active monitoring role.

  • Ofek (1993) finds that the probability of top management turnover following a large stock price decline is inversely related to the ratio of public to total debt.

  • Marshall et al. (2014) document that bank monitoring can play a crucial role in managerial discipline and find a stronger relation between forced CEO turnover and cash flow performance.

  • Industries are controlled for using dummies.

    • The different industries are distinguished based on the WSE industry group definitions.

  • Appendix Table 1A shows the distribution of the sample across industries.

    • The final sample includes 380 public companies with 3,507 observations for each variable for 2000–2015.

CEO Turnover and Founders

  • Volpin (2002) is followed and a probit regression is employed to examine the sensitivity of CEO turnover to firm performance.

  • Dummy variables are used that control for the founder’s position within the firm in terms of both additive effects and interactive effects on the performance measure.

  • The main regression takes the following form:

    • PR(CEO)=α+β<em>1P</em>i,t+β<em>2F</em>i,t+β<em>3C</em>i,t+γ<em>t+δ</em>iP R(CEO) = \alpha + \beta<em>1P</em>{i,t} + \beta<em>2F</em>{i,t} + \beta<em>3C</em>{i,t} + \gamma<em>t + \delta</em>i

    • where CEO is equal to 1 if the firm experiences a forced CEO turnover in firm i and year t, P refers to one of our performance measures, F is a vector representing founders and governance characteristics of the company, and C is a vector representing firm control. γ\gamma represents year fixed effects that capture time-varying macroeconomic trends while δ\delta represents industry fixed effects that control for industry characteristics.

  • All specifications are estimated with robust standard errors clustered at the firm level.

  • Tables 2 to 3 present both the coefficients and marginal effect.

  • Tables 4 to 8 show only the marginal effects to save space and simplify the economic interpretation.

  • The reported marginal effects represent the change in probability of an infinitesimal change in each independent variable evaluated at the average values of the regressors.

  • Firm effects cannot be controlled for in a probit model.

    • All regressions were re-estimated using a linear OLS model with fixed effects at the firm level and robust standard errors.

    • The results of the OLS regressions are in line with the results reported in the study.

    • The results of the OLS regression are available upon request.

Results

CEO Turnover and Firm Performance

  • Table 2 shows how firm performance affects CEO turnover decisions.

  • There should be a greater likelihood that a CEO will be replaced after poor performance.

  • In column (1), the performance measure ROA is contemporary.

  • In column (2), it is lagged by one period.

  • Company performance measured by ROA has a significant effect on the CEO dismissal decision.

  • In columns (1) and (2), the coefficient for ROA is negative and statistically significant at the 1% and 10% levels.

  • When ROA decreases by one standard deviation, the probability of CEO turnover in the current year increases by 4.7%.

  • One year later, the probability of turnover increases only by 1.5%.

  • In columns (3) and (4), the regressions are repeated and market return is employed as the performance measure.

  • CEO turnover does not always decrease with performance when it is proxied by the stock market return.

  • Specifically, in column (3), the coefficient is negative and statistically significant at the 5% level.

  • The probability of CEO turnover in the current year increases by more than 2% if the market return decreases by one standard deviation.

  • In the last columns, the coefficient for market performance is positive and insignificant.

  • The results are consistent with Volpin (2002), who also did not find significant results when performance was proxied for stock return.

  • The coefficient of the leverage variable is positively related to CEO turnover, yet it is significant only in three of the four specifications.

  • This may indicate a weak debtor monitoring mechanism.

  • Marshall et al. (2014) argue that the strength of bank monitoring of poorly performing managers is conditional on firms’ raising of bank debt.

  • CEO succession decisions are unrelated to existing banking relationships, which may explain the relatively weak results.

  • There is also a weak relationship between the ratio of the market to book value of the firm and CEO turnover.

    • This can be explained by the fact that stock value is not the best measure in a shallow market.

  • The control variables are consistent in all the following specifications, so they are not discussed further.

  • Overall, the results confirm that CEO turnover is negatively related to performance.

  • The relationship is significantly weaker when stock return is used as the performance measure.

  • The results are not surprising, as Warner et al. (1988) document that only extreme levels of stock price performance affect the likelihood of top management change.

  • Volpin (2002) argues that stock return is not the best measure of performance in a sample of Italian firms.

    • Stock suffers from a lack of liquidity and infrequent trades.

    • The same argument could be applied to this sample, as the WSE is still a developing and shallow equity market.

  • The stronger effect of contemporaneous financial results on CEO turnover can be explained by the governance of companies in this sample.

  • In Poland, all domestic companies have a two-tier board system.

    • A supervisory board represents all the shareholders who have elected the board to promote their interests.

    • The supervisory board’s main obligation toward the shareholders is to supervise the executive directors.

    • The board has the right to replace the executive directors, including the CEO, at any time.

  • The companies in this sample are listed and obliged to disclose their financial results quarterly.

  • A decline in a company’s financial performance will put pressure on the supervisory board to undertake action, including a change in CEO.

  • If a supervisory board does not undertake any change, its members may be revoked at an extraordinary shareholders general meeting called by one of the block shareholders or at the annual shareholders meeting.

  • There is strong pressure on the supervisory board.

  • In further analysis, only the current financial results of the companies are used in the following regressions.

  • As a robustness check, the lagged variables were also used.

  • Using lagged values, the coefficients do not change the signs, but the results are statistically weaker.

Founder CEO Turnover

  • Table 3 shows whether CEO turnover sensitivities differ if the CEO is also the founder of the company that they manage.

  • The previous estimation was repeated.

    • The dependent variable represents the forced turnover of the founder-CEO.

  • In all the specifications, the performance variables are negative yet statistically significant only when market returns are used as the performance measure.

  • Founder-CEO turnover is only sensitive to market return.

  • The marginal effects in Table 3 are significantly smaller than in Table 2.

    • The likelihood that a founder-CEO will be fired is less likely than an outsider will be fired even when the company under-performs.

  • These results contrast with Jenter and Kanaan (2015), who examine whether CEO-founders are affected differently by peer group performance than other CEOs.

    • The authors find no consistent effects of CEO power on a firm’s propensity to use relative evaluation in their CEO turnover decisions.

  • These results confirm that firm founders hold a special position within a company.

CEO Turnover and Ownership

  • According to entrenchment theory, managers that hold little equity capital in a firm may deploy corporate actions to obtain personal benefits if the shareholders are too dispersed to take action against non-value maximiza- tion behavior.

  • Shleifer and Vishny (1989) develop a model of managerial en- trenchment whereby managers may make specific investments that increase their value to shareholders.

  • Those investments might be aimed at reducing the likelihood of being replaced and may only have entrenchment value.

  • Weisbach (1988) shows that insider ownership is strongly correlated with the composition of the board

    • CEOs have an incentive to avoid including outsiders on the board.

  • Following Jensen and Meckling (1976), Weisbach (1988) argues that as a CEO’s shareholding grows as a fraction of his wealth, the interests of the CEO are more aligned with those of the shareholders.

    • If the CEO has a large equity stake in the firm, there may be less need for monitoring by outside directors.

  • Weisbach (1988) main finding is that outsider-dominated boards are more likely to fire a CEO following poor performance than insider-dominated boards.

    • Outside dominated boards tend to increase shareholder value through their CEO changes.

  • Weisbach (1988) finds that the increased shareholding of a CEO reduces the probability that the CEO will resign, yet the results were statistically insignificant.

  • Insider shareholding may prove an obstacle to those who wish to hasten the departure of the CEO for poor performance.

  • Entrenchment theory is tested by controlling for ownership of the company by a CEO using a dummy variable.

  • Columns (1) and (3) in Table 4 show the results for all the forced CEO turnovers.

  • Columns (2) and (4) show the results for the founder-CEOs.

  • Adding the additional control variable does not change the main results.

    • CEO turnover, including the dismissal of the founder-CEO, is strongly related to firm performance.

    • The marginal effects are, however, significantly smaller for the founder-CEO than for the outsider CEO.

    • In column (2), the coefficient for performance measured by ROA is statistically insignificant.

  • There are significant differences in the impact of outsider CEO and founder-CEO ownership on their dismissals.

  • In columns (1) and (3) of Table 4, the coefficients of variable CEO ownership are negative and statistically significant at the 1% level.

    • A CEO dismissal is less likely if the outsider CEO is a shareholder, which is in line with en- trenchment theory.

  • The interaction term between the performance variable and CEO ownership in columns (1) and (3) shows that the likelihood that a CEO will be dismissed following a bad performance increases if the CEO does not own shares.

    • The results show that the likelihood that an outsider CEO will be fired following bad performance declines if the outsider CEO is also a shareholder of the firm.

    • This situation is attributed to the relatively low governance standards and the lack of shareholder activism in Poland.

  • In contrast, in columns (2) and (4) of Table 4, the coefficients of the variable founder-CEO ownership are positive but statistically significant only in the first specification at the 5% level.

    • Founders who are shareholders are more likely to support their own dismissal as CEO.

    • Founders have a personal attachment to the company they build, which may explain their different behavior compared to outsider CEOs.

  • The results support the arguments of Jensen and Meckling (1976) where the interests of the CEO may align with those of the shareholders.

    • Different behavior is attributed more to personal motives.

  • The interaction term between performance and founder-CEO ownership supports these arguments.

  • In columns (2) and (4) of Table 4, the coefficient of the interaction term is negative and statistically significant in the first specifi- cation at the 10% level.

    • Founder-CEOs face pressure when the company underperforms and, at some point, decide to step down even when they are shareholders.

    • A significant fraction of founders’ wealth is attached to the company, which could also explain their decision to resign as CEO.

  • Whether this ultimately is the driver is unable to be established and left for further research.

  • Insider ownership is related to the dismissal of outsider CEOs and founder-CEOs in different ways.

CEO Turnover and Founders as Monitors

  • Baseline results have been established.

  • The analysis is expanded by controlling for different functions of a founder within the company.

  • By regressing the dependent variable against the performance measures and progressively employing dummy variables that control for the different posi- tions held by a founder within a company.

  • Additionally, in the regressions, a dummy variable is used that controls for insider ownership

  • This analysis allows verification of the previous results on CEO turnover.

  • The analysis allows exploration of situations where the founder is a member of the executive board.

    • Founders voluntarily stepped down as CEO and took a passive role on the board.

    • Doing so gives the founder the advantage of less responsibility while, at the same time, retaining the ability to advise and directly monitor the cur- rent CEO.

    • The founders voluntarily stepped down as CEOs and took a passive role on the board.

  • Investigate the role of founder in the supervisory board where their main responsibility is to monitor the top executive board members.

    • Founders have firm and industry expertise that should allow them to effectively monitor the executive board members.

  • The founder monitors a CEO that the founder chose as their successor and who only follows the founder’s previously outlined strategy for the firm.

  • A founder who assumes a passive role may impede necessary changes in the firm, particularly in the context of a CEO turnover following poor firm performance.

  • The results in Table 5 are in line with previous findings and confirm that a company’s performance measured by ROA strongly affects the dismissal de- cision of the CEO.

  • Controlling for insider ownership and adding the different control variables for the position of the founder within the company hardly changes the coefficient of the performance measure ROA.

    • Remains neg- ative and statistically significant at the 1% level for all the specifications.

  • Insider ownership is negatively related to CEO turnover.

  • The coefficient of the variable proxying for insider ownership is significant in all specifications at least at the 1% level.

  • In columns (1) and (2), a variable is introduced that controls for founder position as CEO and executive board member, respectively.

  • The coefficients of both dummy variables are negative and statistically significant at the 1% level.

  • Founder presence on the executive board influences the CEO dismissal decision.

  • If the founder is a CEO or executive board member, the likelihood that the CEO will be fired declines by 10% and 7%, respectively.

  • Those results are related to the fact that a founder who is a member of the executive board has an impact on the election of a dismissed CEO.

  • The results supplement those of Evans III et al. (2010), who find that former CEOs on the board are likely to have significantly more bargaining power than an exiting CEO as the former CEO is more likely to have been board chairman and/or the firm’s founder.

  • Boards with former CEOs are more likely to select a new CEO who is a younger insider without prior CEO experience and may have family ties to the former CEO.

  • The former CEO is likely to exert significant influence over the successor.

  • A founder on the board is likely to protect a poor-performing CEO as the founder probably selected the CEO and influences his decisions.

  • The interaction terms between the performance measure and the variables for the founder’s position with the executive board align with the assumption.

  • A CEO outsider is likely to be fired, even when the founder is an executive board member, when a company under-performs.

  • Conversely, the situation does not hold if the CEO is the founder, which is in line with the results in 3.

  • Dismissing a founder-CEO can be much more difficult than dismissing an outsider even when the company performs poorly.

  • In columns (3) and (4), a dummy is employed that controls for founder position as chairman and member of the supervisory board.

  • The coefficients of both control variables are statistical insignificant, whereas only the coefficient of the variable controlling for the position of the founder as chairman is negative.

  • There is a weak relationship between the founder supervisory position and CEO turnover.

  • The interaction term that controls for the position of founder as chairman of the supervisory board indicates that the founder may oppose a CEO turnover.

  • There is such a relationship when the founder is only a member of the supervisory board.

  • A founder, who is only a member of the supervisory board, does not have substantial influence on the management of the company any more.

  • The founder role is more likely to be one of adviser or honorary member.

  • The power of a founder to influence CEO turnover declines depending on the founder’s function within the firm.

  • Table 6 shows the results of estimation where the performance variable is the firm’s stock market return.

  • The relationship between turnover and stock return is much weaker than the relationship for ROA.

  • The coefficient of stock market return is negative in all the specifica- tions but statistically insignificant.

  • The coefficient of the variable proxying insider ownership is negative and statistically significant at least at the 5% level.

  • Using stock market return as the performance measure does not change the sign or magnitude of the coefficients of the variables controlling for the founder position as CEO or executive board member, respectively.

  • There is a negligible effect on the decision to fire the CEO.

  • The coefficients for the founder-CEO and founder executive board member are negative and statistically significant at the 1% and 5% levels, respectively.

  • The coefficients of the variable controlling for the position of founder as chairman or as su- pervisory board member are negative, and the first is statistically significant at the 5% level.

  • A founder chairman influences the decision to dismiss an outsider CEO.

  • It contrasts with (Lausten, 2002) who shows that the likelihood of CEO turnover increases following a bad year if the chairman of the board is an insider within the firm.

  • If the founder is a member of the supervisory board, there is a negligible effect on the decision to fire the CEO.

  • The coefficient is insignificant and the marginal effects are close to zero.

  • When controlling for the position of founder within the firm, the relationship between turnover and performance measured by the stock market return is weak as shown by the statistically insignificant in- teraction terms.

  • This finding is attributed to the weak relationship between stock market return and CEO turnover as shown in Table 2.

CEO Turnover and Transition to the Supervisory Board

  • In two-tiered board governance systems, it is not uncommon for CEOs to become a member of the supervisory board.

  • The former CEO has unique expertise that qualifies them as an efficient monitor of manage- ment.

  • The CEO will monitor the decisions of the executive board, which he probably created and recently left.

  • The transition of a former CEO to the supervisory board is a controversial topic in the governance literature (Gerner-Beuerle, 2017).

  • Many countries introduced a cooling-off period for former top executive members’ as a component of their corporate governance codes of best practice.

  • The cooling-off period mitigates a potential conflict of interests between former and current CEOs.

  • In practice, the decision is made by the shareholders who can ignore best prac- tice recommendations emerging during the cooling-off periods.

  • Former CEOs are often elected as members of the supervisory board.

  • Former CEO’s are also elected as chairman of the supervisory board by the other supervisory board members – close to 50% in our sample.

  • The chairman has a relatively strong position on the supervisory board as the chairman calls the meeting and sets up the agenda.

  • Bresser and Thiele (2008) find that, in Germany, a firm’s current CEO is more likely to be dismissed following poor performance when a former CEO is the chairman of the supervisory board.

  • Results show a strong relationship between the position of founder chairman and CEO turnover.

  • Table 7 shows the results where the dependent variable takes the value of one if the CEO turnover results with a transmission of the CEO to the firm’s su- pervisory board.

  • CEO turnover connected with the exiting CEO’s transmission to the supervisory board is not strongly related to company performance.

  • Table 7 shows that the co- efficients of the performance variable ROA are negative yet statistically in- significant in all the specifications.

  • Past performance of the company is not the main determinant for the decision to dismiss the CEO and elect them to the supervisory board.

  • Results supplement the findings of Brickley et al. (1999), who show that the better a firm performs in the years prior to the CEO’s departure, the more likely it is that the former CEO will be offered board positions both in his own firm and in other firms.

  • Brickley et al. (1999) argue that poorly performing CEOs are not only more likely to lose their jobs, they are also less likely to hold board seats after they leave office.

  • Transmission is less likely when the fired CEO is founder of the company.

  • The coefficient of the variable for founder-CEO is negative and statistically significant at the 5% level.

  • The interaction term between the variables F-CEO and the performance variable is statistically significant at the 5% level.

  • Removing a founder of a company might be difficult if the founder is the CEO.

  • If the founder-CEO is fired, they are also less likely to be offered a transition to the supervisory board.

  • The coefficient for insider shareholding is negative yet statistically insignificant in all the specifications.

  • Insider shareholding does not determine CEO transmission to the supervisory board.

  • Similar results are found when stock market return is used as the performance variable.

  • Table 8 shows the results.

  • The coefficient for the performance variable is negative yet insignificant in all the specifications.

  • The only significant coefficient is for the variable for founder-CEO, which is again negative, as well as the interaction term between the variable and the performance variable.

  • Results confirm the special role of the founder within the company.

Robustness

  • The robustness of the main results was checked by conducting a wide array of additional analyses.

  • The first consistency of the results was checked after dropping from the sample all the firms from the financial industry.

    • Dropping the financial firms from the sample does not affect either the significance level or the sign of the estimated coefficients.

  • Alternative performance measures were employed as return on equity, return on assets calculated as earnings before interests and taxes (EBIT) to total assets, or year-to-year changes in EBIT.

  • Using the alternative dependent variables, there is still a significant and neg- ative relationship between firm financial performance and CEO turnover.

  • All the dependent variables were used as lagged by one period in all the specifications.

  • The results for firm performance and the control variables are statistically weaker as shown in Table 2, yet the coefficients do not change the signs and their economic meaning.

  • The results of the robustness test using different methodology, data, and variables confirm the findings concerning the strong link between CEO turnover and firm performance controlling for the different positions of the founder within the firm.

  • The empirical analysis has its limitations

    • Does not determine who can fire a founder-CEO when the firm is underperforming.

Discussion and Conclusions

  • There is considerable interest in the role of founders in the development of their companies, particularly when the founder-CEO is forced to resign.

  • The explores the role of a founder within a firm and their influ- ence on outsider CEO turnover.

    • Uses data from a developing economy, characterized by a low level of governance standards.

    • The environment differs significantly from existing studies on CEO turnover, which is seen as an important internal governance mechanism.

  • There is a negative relationship between CEO turnover and firm performance given the accounting return in the current year.

  • The results are weaker, however, when the market return is used as a performance indicator, in line